Skip to content
codai logo

codai

Routes OpenAI-compatible requests to a Claude model tier by difficulty, checks outcomes, and bills only verified successful tasks.
AgentsFree tier
codai.ro ↗GitHub
AlternativesSet it up for my teamIs codai yours? Claim the card →
At a glanceswept Oct 14, 2026

1.6k installs/mo

Established

For
Developers and teams using Cursor, Claude Code, Copilot or Codex CLI for agentic coding work
Price
From ≈ $5.60/mo€5 · pricing page, read Oct 16, 2026Trial & discounts: 7-day free trial
Activity
Actively maintainedrelease 5 days ago · site updated 4 days ago
Runs as
APISDK
Used for
  • route coding tasks to the matching Claude model tier
  • pay only for tasks verified as successful
  • attach MCP tool servers once for every client
read off codai.ro on Oct 16, 2026
Worth knowingcomputed from what we read · each with its source
  • Has an MCP server (1 in the registry)public registry →
  • Publishes llms.txt, a file written for AI assistantscodai.ro ↗
Pricingpricing page, read Oct 16, 2026

From ≈ $5.60/mo

€5

Trial & discounts 7-day free trial

pricing page

Your datawhat they publish, each line with its page · read Oct 15, 2026

“Your prompts and model outputs are never used to train codai models unless you switch on "Use my data to improve codai".”

security page ↗

These are codai's own statements, read from its pages. We did not audit them. A line that is missing was not found on the pages we read; it does not mean the practice is absent.

How a program reaches itfound, not claimed · read Oct 15, 2026
  • MCP serverro.codai/codaiverified in the MCP registry
  • Stateanswers, requires sign-inchecked 2026-10-15
  • Address to connectai.codai.ro/mcp
  • API docsdocs.codai.ro/apithe vendor's page ↗
  • Status pagelinked from the homepagecodai.ro/status ↗

MCP server details →

Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.

Alternatives in Agentsby use, among tools still maintained

Figure and bar: visits/mo, estimate · year: online since

Compare codai with

Pick up to two.

All Agents tools →

From people who use itnot part of the reading

No notes yet. What do you use codai for?

Sign in to leave a note
Questions this page answers
  • What is codai?

    Routes OpenAI-compatible requests to a Claude model tier by difficulty, checks outcomes, and bills only verified successful tasks. It is listed in the Atlas under Agents.

  • Who is codai for?

    Developers and teams using Cursor, Claude Code, Copilot or Codex CLI for agentic coding work

  • How much does codai cost?

    Paid plans start at ≈ $5.60/mo (€5/mo), per codai's pricing page on 2026-10-16.

  • Do people use codai?

    Established: 1.6k installs/mo (npm).

  • Is codai still maintained?

    Actively maintained: release 5 days ago · site updated 4 days ago.

  • Does codai train AI on your data?

    Only with your consent or on some plans, per its security page (2026-10-15): “Your prompts and model outputs are never used to train codai models unless you switch on "Use my data to improve codai".”

  • Is codai safe?

    We do not audit vendors. What codai publishes, read 2026-10-15: GDPR stated on its privacy policy; on training with your data, “Your prompts and model outputs are never used to train codai models unless you switch on "Use my data to improve codai".”.

  • Does codai have a free trial?

    Yes, a 7-day free trial, per its pricing page (2026-10-15).

  • What are alternatives to codai?

    In Agents, by use: Open Agent Skill, Cuey, CalendarMCP, Dijin, /mission for Claude Code, Meltbox.

Not quite it? Describe the job

beta

Five tools from the Atlas that fit it, each judged against what you wrote.

How we read a tool

No votes, no reviews, no vendor claims. Every week a crawler reads each tool's own site and a few public registries, and the words on the card are bands over what it read.

  • Use: visits to the site (estimated), installs from npm and PyPI, presence in Chrome's usage report.
  • Activity: the newest release on GitHub, npm or PyPI; the newest dated page on the site; open roles on a public jobs board.
  • Price: the vendor's own pricing page, read with the date. A figure is printed only when it is on that page.
  • The line and the use cases are written by us from the homepage, one tool at a time, and refused when they repeat the vendor's marketing.

How AI assistants read each site — the reading vendors ask us about — is on each tool's own visibility page.